Estimation in medical imaging without a gold standard

Matthew A Kupinski, John W. Hoppin, Eric W Clarkson, Harrison H Barrett, George A. Kastis

Research output: Contribution to journalArticle

27 Citations (Scopus)

Abstract

Rationale and Objectives. In medical imaging, physicians often estimate a parameter of interest (eg, cardiac ejection fraction) for a patient to assist in establishing a diagnosis. Many different estimation methods may exist, but rarely can one be considered a gold standard. Therefore, evaluation and comparison of different estimation methods are difficult. The purpose of this study was to examine a method of evaluating different estimation methods without use of a gold standard. Materials and Methods. This method is equivalent to fitting regression lines without the x axis. To use this method, multiple estimates of the clinical parameter of interest for each patient of a given population were needed. The authors assumed the statistical distribution for the true values of the clinical parameter of interest was a member of a given family of parameterized distributions. Furthermore, they assumed a statistical model relating the clinical parameter to the estimates of its value. Using these assumptions and observed data, they estimated the model parameters and the parameters characterizing the distribution of the clinical parameter. Results. The authors applied the method to simulated cardiac ejection fraction data with varying numbers of patients, numbers of modalities, and levels of noise. They also tested the method on both linear and nonlinear models and characterized the performance of this method compared to that of conventional regression analysis by using x-axis information. Results indicate that the method follows trends similar to that of conventional regression analysis as patients and noise vary, although conventional regression analysis outperforms the method presented because it uses the gold standard which the authors assume is unavailable. Conclusion. The method accurately estimates model parameters. These estimates can be used to rank the systems for a given estimation task.

Original languageEnglish (US)
Pages (from-to)290-297
Number of pages8
JournalAcademic Radiology
Volume9
Issue number3
DOIs
StatePublished - 2002

Fingerprint

Diagnostic Imaging
Regression Analysis
Noise
Statistical Distributions
Hospital Distribution Systems
Nonlinear Dynamics
Statistical Models
Linear Models

Keywords

  • Estimation
  • Gold standard
  • Image-quality assessment
  • Maximum likelihood

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging

Cite this

Estimation in medical imaging without a gold standard. / Kupinski, Matthew A; Hoppin, John W.; Clarkson, Eric W; Barrett, Harrison H; Kastis, George A.

In: Academic Radiology, Vol. 9, No. 3, 2002, p. 290-297.

Research output: Contribution to journalArticle

@article{d24a9ee5e1e748008db20bd9b80271bf,
title = "Estimation in medical imaging without a gold standard",
abstract = "Rationale and Objectives. In medical imaging, physicians often estimate a parameter of interest (eg, cardiac ejection fraction) for a patient to assist in establishing a diagnosis. Many different estimation methods may exist, but rarely can one be considered a gold standard. Therefore, evaluation and comparison of different estimation methods are difficult. The purpose of this study was to examine a method of evaluating different estimation methods without use of a gold standard. Materials and Methods. This method is equivalent to fitting regression lines without the x axis. To use this method, multiple estimates of the clinical parameter of interest for each patient of a given population were needed. The authors assumed the statistical distribution for the true values of the clinical parameter of interest was a member of a given family of parameterized distributions. Furthermore, they assumed a statistical model relating the clinical parameter to the estimates of its value. Using these assumptions and observed data, they estimated the model parameters and the parameters characterizing the distribution of the clinical parameter. Results. The authors applied the method to simulated cardiac ejection fraction data with varying numbers of patients, numbers of modalities, and levels of noise. They also tested the method on both linear and nonlinear models and characterized the performance of this method compared to that of conventional regression analysis by using x-axis information. Results indicate that the method follows trends similar to that of conventional regression analysis as patients and noise vary, although conventional regression analysis outperforms the method presented because it uses the gold standard which the authors assume is unavailable. Conclusion. The method accurately estimates model parameters. These estimates can be used to rank the systems for a given estimation task.",
keywords = "Estimation, Gold standard, Image-quality assessment, Maximum likelihood",
author = "Kupinski, {Matthew A} and Hoppin, {John W.} and Clarkson, {Eric W} and Barrett, {Harrison H} and Kastis, {George A.}",
year = "2002",
doi = "10.1016/S1076-6332(03)80372-0",
language = "English (US)",
volume = "9",
pages = "290--297",
journal = "Academic Radiology",
issn = "1076-6332",
publisher = "Elsevier USA",
number = "3",

}

TY - JOUR

T1 - Estimation in medical imaging without a gold standard

AU - Kupinski, Matthew A

AU - Hoppin, John W.

AU - Clarkson, Eric W

AU - Barrett, Harrison H

AU - Kastis, George A.

PY - 2002

Y1 - 2002

N2 - Rationale and Objectives. In medical imaging, physicians often estimate a parameter of interest (eg, cardiac ejection fraction) for a patient to assist in establishing a diagnosis. Many different estimation methods may exist, but rarely can one be considered a gold standard. Therefore, evaluation and comparison of different estimation methods are difficult. The purpose of this study was to examine a method of evaluating different estimation methods without use of a gold standard. Materials and Methods. This method is equivalent to fitting regression lines without the x axis. To use this method, multiple estimates of the clinical parameter of interest for each patient of a given population were needed. The authors assumed the statistical distribution for the true values of the clinical parameter of interest was a member of a given family of parameterized distributions. Furthermore, they assumed a statistical model relating the clinical parameter to the estimates of its value. Using these assumptions and observed data, they estimated the model parameters and the parameters characterizing the distribution of the clinical parameter. Results. The authors applied the method to simulated cardiac ejection fraction data with varying numbers of patients, numbers of modalities, and levels of noise. They also tested the method on both linear and nonlinear models and characterized the performance of this method compared to that of conventional regression analysis by using x-axis information. Results indicate that the method follows trends similar to that of conventional regression analysis as patients and noise vary, although conventional regression analysis outperforms the method presented because it uses the gold standard which the authors assume is unavailable. Conclusion. The method accurately estimates model parameters. These estimates can be used to rank the systems for a given estimation task.

AB - Rationale and Objectives. In medical imaging, physicians often estimate a parameter of interest (eg, cardiac ejection fraction) for a patient to assist in establishing a diagnosis. Many different estimation methods may exist, but rarely can one be considered a gold standard. Therefore, evaluation and comparison of different estimation methods are difficult. The purpose of this study was to examine a method of evaluating different estimation methods without use of a gold standard. Materials and Methods. This method is equivalent to fitting regression lines without the x axis. To use this method, multiple estimates of the clinical parameter of interest for each patient of a given population were needed. The authors assumed the statistical distribution for the true values of the clinical parameter of interest was a member of a given family of parameterized distributions. Furthermore, they assumed a statistical model relating the clinical parameter to the estimates of its value. Using these assumptions and observed data, they estimated the model parameters and the parameters characterizing the distribution of the clinical parameter. Results. The authors applied the method to simulated cardiac ejection fraction data with varying numbers of patients, numbers of modalities, and levels of noise. They also tested the method on both linear and nonlinear models and characterized the performance of this method compared to that of conventional regression analysis by using x-axis information. Results indicate that the method follows trends similar to that of conventional regression analysis as patients and noise vary, although conventional regression analysis outperforms the method presented because it uses the gold standard which the authors assume is unavailable. Conclusion. The method accurately estimates model parameters. These estimates can be used to rank the systems for a given estimation task.

KW - Estimation

KW - Gold standard

KW - Image-quality assessment

KW - Maximum likelihood

UR - http://www.scopus.com/inward/record.url?scp=0036183057&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0036183057&partnerID=8YFLogxK

U2 - 10.1016/S1076-6332(03)80372-0

DO - 10.1016/S1076-6332(03)80372-0

M3 - Article

C2 - 11887945

AN - SCOPUS:0036183057

VL - 9

SP - 290

EP - 297

JO - Academic Radiology

JF - Academic Radiology

SN - 1076-6332

IS - 3

ER -